In this proposal, we seek to understand RNA dynamics in transient cellular complexes and to use that information to identify candidate drugs that may modulate the pathways to malignancy. While protein components of RNP granules have been studied in mammalian cells, quantitative RNA population dynamics have not been analyzed for lack of suitable technologies. Our laboratory focuses on multi-targeting of populations of mRNAs regulated by RNA-binding proteins (RBPs) and microRNAs. We have demonstrated that sequence specific RBPs coordinately regulate groups of functionally related mRNAs and these RNPs are remodeled during activation of cells with small molecule drugs. For example, the mRNAs associated with several RBPs have been shown to change coordinately following treatment of embryonic carcinoma cells with retinoic acid or leukemia-derived immune T cells with phorbol esters plus mitogens. We hypothesize that dynamic changes in RNA populations within these transient RNP complexes coordinate functionally-related subsets of mRNAs that encode proteins whose synchronized expression is required for oncogenic transformation. Here, we will use a probabilistic approach that quantifies dynamic RNA changes en masse to analyze mRNAs associated with RBPs during progression to malignancy. We will examine the transition from a precancerous state to a cancerous state in primary epithelial cells using methods pioneered in the laboratory of Robert Weinberg with RAS, telomerase and four other transforming proteins quantify dynamic changes in RNAs associated with HuR (early response gene mRNAs), TIAR (stress granule RNAs) and AGO2/RISC (processing P body RNAs). HuR shift to the cytoplasm alters its mRNA targeting and has been claimed as a prognostic factor in hereditary breast cancer. We will quantify the levels of mRNAs and microRNAs in these complexes and determine how they change in response to progressive development of a transformed phenotype, as well as after inducing oxidative stress and hypoxia. The RNP-Immunoprecipitation microarray (RIP-Chip) procedure will be used to identify and quantify mRNAs associated with specific RBPs; an ultraviolet light crosslinking procedure with high specificity and efficiency termed PAR-CLIP recently developed in the laboratory of our collaborator, Thomas Tuschl, will be used to identify the precise binding sites of these RBPs and microRNAs. Dynamic changes in sites of microRNA binding to mRNAs will be globally integrated with RBP binding sites. We will construct a quantitative dynamic model of these events and use these data to query the drug-genome Connectivity Map for drugs that affect these processes as demonstrated in our recent publications. These compounds will be used to further investigate the underlying biology of carcinogenesis in this system. Microscopic visualization at each stage of progression will be used to confirm the RNA/RBP localization and phenotypic effects of drug treatments. It is our long-term plan to use this quantitative probabilistic approach of RNA targeting to investigate these and other transient macromolecular RNP complexes involved in posttranscriptional gene expression using animal models of cancer.